By Thomas Kaufman (follow me on Twitter)

Duran v. U.S. Bank, is notable because it is the first decision to analyze thoroughly the defendant’s due process rights as they were handled in one of the "innovative" class trial procedures that Sav-On v. Superior Court encouraged trial courts to formulate.  Before this case, the only case that significantly addressed class trial procedure was Bell v. Farmers InsuranceBell, however, involved only a trial on damages after a court held that the defendant had misclassified all of its insurance adjusters as exempt.  Because liability was already decided classwide, the only issue was how much of a recovery each class member was entitled to receive.  What is worse, the Bell defense counsel waived several defenses by attempting to be "cooperative" with opposing counsel and thereby could not assert several good arguments on appeal.  Much mischief has been made by courts since Bell applying it as some sort of a template on how to conduct a class trial on liability.

Duran is strikingly different because U.S. Bank was effectively dragged kicking and screaming to trial, and it repeatedly objected to the many "innovative" procedures the trial court implemented.  Accordingly, the case presented the court of appeal with numerous, solid examples of a trial court running roughshod over the defendant’s due process rights in the spirit of attempting to formulate a "streamlined" trial procedure.  The case thus provides binding authority (assuming the California Supreme Court does not grant review) that employers can cite when arguing that the plaintiff’s trial plan improperly deprives the defendant of due process. In fact, if the guidance of this decision is followed, it is hard to see how many wage hour class actions that are routinely certified could actually proceed to trial.

The Basic Facts

U.S. Bank employed roughly 260 Business Banking Officers (BBOs) in California during the relevant class period.  The evidence appeared to be largely undisputed that BBO was a sales job in which the employees follow leads for small business banking customers to whom they would attempt to sell financial services products such as credit, deposit, cash management, and other bank products and services.  Although U.S. Bank argued that the commission sales exemption, administrative exemption, and outside sales exemption applied to this job, the court granted summary adjudication to the plaintiff on the commission sales and administrative exemption, leaving only the outside sales exemption in dispute.  The dispute over the outside sales exemption appeared to be limited to whether the BBOs spent the majority of their time outside as opposed to working on U.S. Bank property. 


At class certification, each party submitted a substantial number of declarations in which competing groups of BBOs attested that they either worked inside or outside (similar to Vinole v. Countrywide, which is cited favorably in the case).  Rather than conclude that whether BBOs spent sufficient time outside to meet the outside sales exemption was a predominant individualized issue, Judge Robert Freedman of the Alameda complex court instead certified the class and order a class "Trial by Formula" where a small, purportedly "representative sample" of the class would present their claims to the finder of fact, and the determination of whether that sample was exempt would be extrapolated to the larger class. 


The plaintiffs waived jury so the case was presented as a bench trial.  Using his own brand of statistical analysis, Judge Freedman decided that 20 class members would be selected at random from the class list (opt outs would not be considered) and that they and the two plaintiffs would serve as the test group to evaluate the exemption.  To make matters worse, after the 20 people were selected, they were given a chance to opt out of the case in which case they would be excused from testifying and "alternates" would take their place.  According to the defendant, this simply caused the people selected to be in the sample who believed they were exempt to opt out of the case and skewed the sample further to the plaintiff.  Furthermore, at trial, Judge Freedman refused to consider testimony from any BBOs except those in the final "sample," going so far as to bar any testimony from other BBOs showing that they spent the majority of their time outside, and thus were exempt.  Defendant proffered 70 BBOs who would testify that they spent the majority of their time outside, but they were barred from testifying at all.


Judge Freedman then conducted the trial and concluded that U.S. Bank could not prevail as to any of the BBOs in the sample because the evidence was that these BBOs were free to spend their time working inside or outside with U.S. Bank caring only about their productivity and not about whether they worked outside enough to meet the exemption.  From that premise, Judge Freedman found that the entire class was misclassified.  Judge Freedman also incorporated the plaintiffs’ statistics expert’s conclusion that the average class member worked 11.87 overtime hours per week.  The expert plaintiff used was Richard Drogin, the same expert the plaintiffs used in Bell v. Farmers Insurance.


Once liability was determined, Judge Freedman held a second phase of the trial where Drogin testified that, with a 95% confidence level, the overtime worked by the average class member was within 5.14 hours of the 11.87 hour figure adopted in phase 1, a margin of error of 43%.  Plaintiffs then had an accounting expert testify that total damages were approximately $14 million if the overtime estimate provided by Drogin was accepted.  The defense put on an expert who opined that using a different methodology, overtime could be calculated at only 6.73 hours per week.  U.S. Bank also put on an expert who challenged the scientific basis for the "sampling" methodology the court had utilized.  The trial court sided with the plaintiffs and awarded approximately $14 million in damages, inclusive of prejudgment interest.  U.S. Bank appealed.


The Many Good Holdings in the Decision


The court of appeal went through every aspect of how Judge Freedman had handled the trial and rejected every significant decision he had made.  The court rejected Judge Freedman’s application of the outside sales exemption, his unscientific sampling methodology, and his utter disregard for U.S. Bank’s right to defend itself as to individual class members.  Here is a list of just the most notable holdings in the case:


(1) The court of appeal provided a definition of due process that a defendant can invoke whenever the class procedure short circuits its right to defend itself:


"Due process principles are designed to ensure a party is afforded his or her right  to be heard during adversarial proceedings.  As the rubric itself implies, procedural due process is simply a guarantee of fair procedure.  Hence we review cases involving adversarial hearings to determine whether, under the specific facts and circumstances of a given situation, the affected individual has a fundamentally fair chance to present his or her side of the story." 


The court of appeal also noted that, even when the California Supreme Court in Sav-On encouraged courts to be innovative with class trial procedures, it stated that the innovative procedures still must "protect[] the rights of all the parties."  As the court later put it: "[W]e have never advocated that the expediency afforded by class action litigation should take precedence over a defendant‘s right to substantive and procedural due process."


(2) The trial court’s statistical sampling methodology was improper because the only disputed material issue was whether individual class members spent the majority of their time "outside" and the methodology deprived defendant of the right to prove that individual class members did so.  Indeed, Judge Freedman found liability as to everyone in the class even though class members themselves testified they spent the majority of their time outside.  Judge Freedman erroneously concluded that the employer did not take steps to force employees to spend the majority of the time outside which meant that it did not sufficiently set expectations that the job had to be carried out in an exempt fashion.  The court of appeal noted that Judge Freedman got this exactly backwards– i.e., the absence of any focus on whether employees spent their time inside or outside likely meant that some BBOs met the exemption and some didn’t.  That should have weighed against class certification, not for it.  (citing Spainhower v. U.S. Bank National Association (C.D.Cal. Mar. 25, 2010)).


(3) The court appeared to reject the use of statistical sampling to determine liability in almost any case where the defendant could show variation among the class as to liability.  The court cited with approval Dukes v. Wal-Mart for the proposition that a class action may not be based on Trial by Formula where a sample of the class is evaluated and liability and damages of the sample are extrapolated to the larger class.  Implicitly rejecting the argument that Dukes only applies under Title VII or in federal courts, the Duran court held that Judge Freedman’s trial procedure was fatally flawed for the same reason the trial in Dukes had been flawed:


"The same type of ‘Trial by Formula’ that the U.S.Supreme Court disapproved of in Wal-Mart is essentially what occurred in this case. It is important to appreciate this portion of the Wal-Mart opinion was the expression of a unanimous court."


Furthermore, at footnote 72, the court appeared to issue a blanket prohibition on using sampling to prove classwide liability because doing so was inconsistent with the United States’ tort system:


"[U]nder current law sampling is a practical option only at the damages stage. There is no conceptual obstacle to using sampling to measure liability, but it would require a major change in tort law. Tort liability is binary: a defendant is either liable or not, and if liable, the defendant must compensate the plaintiff in full. At best, sampling applied to liability can only provide an estimate of the probability that defendant is liable to any plaintiff in an arbitrarily chosen case. This estimate equals the number of liability verdicts divided by the total number of sample cases. Thus, sampling could be used to determine liability only if the tort law recognized probabilistic liability measures."


(4) The court of appeal held that it was a denial of due process to refuse to allow the defendant to present evidence from class members that would establish that those class members were properly classified as exempt.  The court noted that its statement in Bell that the defendant’s due process right was only as to the total amount of damages owed to the class and not as to the distribution of those damages to individual class members had to be understood in the context of the fact that (a) classwide liability had already been determined in that case, (b) "the employer had acquiesced to statistical proof of damages and had waived the right to impeach the employees‘ testimony at trial."  The court seemed to accept that the defendant does have a due process right not to pay money to an employee to whom it has no liability at all.


(5) The court of appeal held that, even assuming statistical sampling were proper in some cases, Judge Freedman’s use of a sample of 20 BBOs to represent 260 had no scientific basis at all.  What’s more, his decision to allow the named plaintiffs to be added to the sample and to allow BBOs selected for the sample to opt out rendered his sampling methodology junk science.  If the result of using a sample large enough to be scientifically reliable is an unmanageable trial, then the court should not certify the case.


(6) The conclusion of the plaintiffs’ own expert that there was a 43% margin of error in the estimate of how much overtime class members worked rendered his conclusions too imprecise to satisfy due process requirements.  The court of appeal noted that even in Bell the court of appeal reversed the double-time award when the evidence showed that there was about a 30% margin of error in its calculation. 


(7) Given the record the trial court was presented, it was error not to decertify the case.  The court of appeal does not pinpoint precisely which aspect of the record warrants decertification, but seemed to hold that if there was ever a case that should have decertified, this was it:

"At this juncture, we need not speculate as to whether a workable trial plan could have been devised to account for these individual inquiries. In view of the many courts that have considered this problem at the classification stage, it is doubtful that such a plan could be successfully implemented. Here, the trial court attempted to manage the individual issues in the first phase of this trial by resorting to an unproven statistical sampling methodology that denied USB the right to properly defend the claims against it. As we have demonstrated, the plan fell short. Accordingly, we conclude the failure to grant USB‘s second motion to decertify was an abuse of discretion."

Why The Case Is Significant


When Dukes v. Wal-Mart came down and appeared to say that Trial by Formula was a violation of constitutional due process, many thought this would render the great majority of class actions uncertifiable.  Except in the smallest cases, or in cases where a plaintiff truly does put a single common practice on trial, a plaintiff’s trial plan almost always relies on some use of statistical sampling akin to what Judge Freedman attempted.  Judge Freedman’s was just an extreme example of a trial procedure that seemed to go out of its way to deprive the defendant’s of a fair day in court.  Other courts certifying classes have approved procedures that differ only in degree, not basic form.


Since Dukes, a number of individual courts have attempted to limit the Supreme Court’s unanimous rejection of Trial by Formula to massive discrimination cases, or cases arising under Title VII, or cases arising in federal court.  Duran cogently explains why the same problems that Supreme Court identified in Dukes apply to actions in California state courts as well.  They confirm that the issues Dukes raised are due process issues, and not issues of particular federal laws.  The fact that Duran was issued by the same appellate district as decided Bell (Marchiano was on the panel that decided both Bell III and Duran) gives this decision added salience, as the Duran court appears to harmonize its decision with Bell rather than create a split of authority. 


Here’s hoping the California Supreme Court declines review.